How to Validate Master Equipment that has been brought to onsite

E

EMacP

#1
Hi All,

My 17025 audit just ended. Currently my laboratory provide onsite calibration for our customers who can't afford downtime on their equipment. Inside my SOP, it is stated that before bringing out the standard, engineer need to perform selftest/selfcal making sure the standard is working, redo the whole process at customer site and then finally when standard return from onsite. One of the auditor ask me whether by performing selftest/selfcal is good enough, how do I know the the calibration values will not be affected when the standard is transported to and fro. My questions are, is it necessary for me to re-validate my standard by performing re-calibration of my standard every time if I took the unit out? Can we sample check at some parameter points? Currently my standards were used for onsite probably 5-10 times in a year. Is there a method to do so?:confused:

A general question on clamp meter, how do we know the clamp meter used is a toroidal type or not? Most of the time there is no indication on the datasheet. I wanted to apply for accreditation for this and when i read on the specs for current coil that I am using, it specified 2 separate specification, one is for toroidal-type clamp meter and the other is any other type.:thanx: for the help.
 
Elsmar Forum Sponsor
D

dv8shane

#2
Hi All,

My 17025 audit just ended. Currently my laboratory provide onsite calibration for our customers who can't afford downtime on their equipment. Inside my SOP, it is stated that before bringing out the standard, engineer need to perform selftest/selfcal making sure the standard is working, redo the whole process at customer site and then finally when standard return from onsite. One of the auditor ask me whether by performing selftest/selfcal is good enough, how do I know the the calibration values will not be affected when the standard is transported to and fro. My questions are, is it necessary for me to re-validate my standard by performing re-calibration of my standard every time if I took the unit out? Can we sample check at some parameter points? Currently my standards were used for onsite probably 5-10 times in a year. Is there a method to do so?:confused:
I suggest you create a form to take readings outgoing and incoming for each of the standards. You do not have to fully calibrate the unit a single reading for each function should do. You just have to be able to prove it did not shift between when it left your lab and when it returned. i.e. Multifunction calibrator test DC at 10 V AC at 10 V 1 kHz 10 k Ohm etc against your reference meter. Decade box set all to maximum etc.

Shane
 
G

George Weiss

#3
Hello new poster,

In agreement with last post, and going for more!

You have 2 questions.
1) What kind of verification do I do for onsite calibration stds?
2) Toroidal coils and all others?
You stumbled into the guy who has gone through this also.
Question #1
The 17025 accreditation audit team will ask you to jump through many hoops. You have to do what you can. They will ask you to protect your standard(s) in extreme ways to avoid them becoming damaged. You mentioned the request by the auditors to do a 4 point test. Before/after going, and before/after coming home. The short and sweet of it is, that a quick minimal reference measurement before leaving, and then when back at the lab is enough. Make a DCV check, and or Ohms test with your 3458A, and expect them to be the same when you return. The amount of change which you should allow is for you to decide. Adding to that a complete self-check or self-cal-check is even better.
The method of testing the standard(s) you take is up to you. The auditor will review your methods and ALWAYS ask for more. Having a good position is as important as a good procedure. In practice the vast majority of equipment going to the field and back is never checked. I would however QUICK-SPOT-CHECK your multifunction calibrators like 5700A or 5500A, and other sensitive items. Every year the auditor will ask for more, and look deeper into your operation. Every year they will expect more. This is how the game is played. Continual improvement via accreditation audit findings.
Question #2
You didn’t say, but I knew you are talking about the Fluke 50 turn coil.
The big question is how do you calibrate this thing? It is an act of god! Likely you cannot.
The good news is, that asking Fluke to do a 17025 calibration would be a one-time deal.
If a auditor sees a 17025 calibration from Fluke 1, or 5, or 10 years ago, then you are covered.
The coil has been written up in a past issue of NCSL, I don’t have the copy in front of me, that reports on this coil. Basically they are saying it is a device that once determined working, will work until broken.
The NCSL article is worth finding on the NCSL website.
The re-calibration of the coil will be like the annual re-calibration of a glass ruler.
Generally you would calibrate a standard at all possible points, but here I am advising 1 point at 60hz at low current. I am not the god on this, but This coil is almost best left as a FNC calibration.
50 turn coil specs @
http://assets.fluke.com/manuals/5500coiliseng0800.pdf
you can really go inside-out trying to calibrate this with some kind of accuracy, or with a TUR better than 1:1
The comments on the NCSL article mentioned, I believe also the location of your test clamp can cause variations.
Basically it works, and the person using the coil will cause potential errors in the process.
 
Last edited by a moderator:
D

dv8shane

#4
I agree with George, but have some more supporting information in regards to the coil I have a description of how to build a 50 turn coil in NAVAIR 17-20AQ-32. Once built it is 50 ± X uncertainty which is good enough for the Naval Surface Warfare Center. If you have access to the GIDEP data base check it out
 
E

EMacP

#5
:agree1:You are right on the coil thing. I have the 50 turn coils. The technical accessor do accept me using the coil specs as one of my uncertainty components. The one I forgot to include is the uncertainty deviation with respect to positioning, eventhough I demonstrate to the accessor that no deviation presence. Still unacceptable. So I need to resubmit computation.

Question#1

Currently I have many standards, both electrical and mechanical equipment that are used for the onsite. To perform 4 time checks will be very tedious. Even more so now that I am required to perform trending for most of my standards. It is really tough for commercial lab like mine which focus on revenues.:nope:
 
G

George Weiss

#6
ISO 17025 presses for better, and at a point the costs push back. The auditors know this. You should improve as the auditor had suggested, and find what is realistic. Document what you can do, and do that. If you document unrealistically, then you will receive an audit finding, that you didn't do as your procedure says. "write what you do", and then, "do as you write". Words to go by.
I would not worry about testing most mechanical items. A gage block set or pin gage set need no re-checking.
A comment about the 50 turn coil: If the auditor picked on this item, then:
1. The auditor knew it was a weakness for any lab, and saw an easy ding. An easy-ding is good for you and the auditing body. You both save face, and the lab is re-accredited.
2. Likely your lab is in pretty good shape if he was asking for re-calculation of the uncertainties for the 50 turn coil. PLEASE! wow....
I found the 50 turn coil article posted on the June 2009 A2LA newsletter. Page #10. Good reading! @
http://www.a2la.org/newsletters/June_2009_Newsletter.pdf
I am sure that the auditor has read this article. So now you will be talking to him on level ground.
If this was your only ding, then have a party ! ! ! !
 
Last edited by a moderator:
E

EMacP

#7
ISO 17025 presses for better, and at a point the costs push back. The auditors know this. You should improve as the auditor had suggested, and find what is realistic. Document what you can do, and do that. If you document unrealistically, then you will receive an audit finding, that you didn't do as your procedure says. "write what you do", and then, "do as you write". Words to go by.
I would not worry about testing most mechanical items. A gage block set or pin gage set need no re-checking.
A comment about the 50 turn coil: If the auditor picked on this item, then:
1. The auditor knew it was a weakness for any lab, and saw an easy ding. An easy-ding is good for you and the auditing body. You both save face, and the lab is re-accredited.
2. Likely your lab is in pretty good shape if he was asking for re-calculation of the uncertainties for the 50 turn coil. PLEASE! wow....
I found the 50 turn coil article posted on the June 2009 A2LA newsletter. Page #10. Good reading! @
http://www.a2la.org/newsletters/June_2009_Newsletter.pdf
I am sure that the auditor has read this article. So now you will be talking to him on level ground.
If this was your only ding, then have a party ! ! ! !
I had read the newsletter and as you can see the uncertainty budget components get more. It does provide a better reflection of the uncertainty measurement capability of the lab. That is also one of the requirement by the technical accessor for my temperature lab to perform stability tracking on reference equipment and include it in the uncertainty budget computation.
 
G

George Weiss

#8
That is another easy-ding from the accreditation auditors, "The temperature uncertainty" concern(s).
The H/P 3458A has a button to determine/monitor various logged temperatures. This meter is focused on normally as a top lab measurement standard. Many pieces of equipment, and artifact standards, have temperature coefficients in their spec-sheets. Your problem(s) are normal for the first few times through accreditation. The problems, and their solutions become a bit more involved in later deficiencies. You will be ready for them by that time.
Locating an audit check list will help in expecting the issues reported.
You are likely a lab supervisor/support engineer, and some planning and reading here is worth the investment.
 
E

EMacP

#9
That is another easy-ding from the accreditation auditors, "The temperature uncertainty" concern(s).
The H/P 3458A has a button to determine/monitor various logged temperatures. This meter is focused on normally as a top lab measurement standard. Many pieces of equipment, and artifact standards, have temperature coefficients in their spec-sheets. Your problem(s) are normal for the first few times through accreditation. The problems, and their solutions become a bit more involved in later deficiencies. You will be ready for them by that time.
Locating an audit check list will help in expecting the issues reported.
You are likely a lab supervisor/support engineer, and some planning and reading here is worth the investment.
You are right on the temperature uncertainty. I am aware on the temperature coefficient for artifact standards. Eventhough the uncertainty variations maybe insignificant to the meter if the uncertainty of the resolution of the uut become big . However the component has to be included within the budget computation. Currently the stability tracking on reference equipment, like the thermocouple, rtd and so on for my temperature side has become a must for my current audit.

Since you mention about 3458A. I thought of creating a new thread but I think I can just asked anyone who can help me on my question. Currently if I want to calibrate a 3458A under 17025, I either need to send my master equipment 5720A or 3458A for 90days cal. What if I don't wish to go for 17025, can I still calibrate 3458A for customer without sending any of my master equipment for 90days cal but normal 12 mths cal interval. The TUR for some of measurement parameters will become 2:1. Do I need to inform customer on this matter?
 
G

George Weiss

#10
There is calibration, and there is verification.
calibration is with 10.00000VDC and 10.00000 kOhms,, and the short. (The DC calibration)
You do not want to do the AC calibration unless you know for sure there is a problem with the AC ranges.
verification is with many standards.
ISO 17025 states that uncertainty should be considered, and needs to be stated.
There are many uncertainty threads on elsmar.
There are some guard banding approach theads also
Agilent has white papers and discussions on guard band approaches.
with 4:1 TUR you are in good shape.
With less you have chances of questionable test results. You can do it. But the fuzzy area grows.
I am not sure about what you are calibrating.
The 3458A can be characterized to meet special testing some times.
Doing a 17025 calibration should not be ruled out.
 
Thread starter Similar threads Forum Replies Date
L How to determine / validate Medical Device Storage Conditions ISO 13485:2016 - Medical Device Quality Management Systems 1
Anonymous16-2 21 CFR Part 11 - Steps to take if we want to validate an electronic system Pharmaceuticals (21 CFR Part 210, 21 CFR Part 211 and related Regulations) 2
T How to “validate/qualify” a new measurement equipment? Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 4
R Validation of Medical Device Hardware containing Software - How many to Validate ISO 13485:2016 - Medical Device Quality Management Systems 1
S Do we need to validate Software used in Drug discovery and development process? Qualification and Validation (including 21 CFR Part 11) 2
kisxena Memo to File - Failure to Validate Supply Chain Security Management Systems 1
A ISO 13485:2016 - Validate our use of software that impacts on the QMS ISO 13485:2016 - Medical Device Quality Management Systems 8
A Process to Validate the Accuracy of Test Reports (AS9100) AS9100, IAQG, NADCAP and Aerospace related Standards and Requirements 38
C Do we need to Validate Off the Shelf Software? Manufacturing and Related Processes 5
M How to validate hundreds or even thousands of parts? Qualification and Validation (including 21 CFR Part 11) 7
B How to validate Microsoft Excel Spreadsheet Calculations for the Cpk value Capability, Accuracy and Stability - Processes, Machines, etc. 9
M How to validate a system to make it compliant with 21 CFR Part 11 Qualification and Validation (including 21 CFR Part 11) 2
B Do I need to validate my database software? Calibration and Metrology Software and Hardware 1
K COTS Validation: Should we validate Windows after each Update? IEC 62304 - Medical Device Software Life Cycle Processes 3
S How to validate a homegrown software 21 CFR Part 820 - US FDA Quality System Regulations (QSR) 5
A How to verify and validate the effectiveness of CAPA with regards to employees ? 21 CFR Part 820 - US FDA Quality System Regulations (QSR) 5
A When to verify and when to validate Design Changes 21 CFR Part 820 - US FDA Quality System Regulations (QSR) 4
H How To Validate Electronic Part From Non Authorized Distributor Process Maps, Process Mapping and Turtle Diagrams 7
G Prototype Manufacturing Equipment - Validate or not to validate? Quality Manager and Management Related Issues 2
S How to validate Device Shelf Life? CE Marking (Conformité Européene) / CB Scheme 8
M Destructive VTMV Subgroups - A Destructive Variable Test Method to Validate Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 11
M How to validate a Dye Test method used to evaluate Package Integrity Pharmaceuticals (21 CFR Part 210, 21 CFR Part 211 and related Regulations) 8
I When to Validate and when not to Validate US Food and Drug Administration (FDA) 24
C How do you Validate Solderability of Components Prior to Soldering Operations Manufacturing and Related Processes 5
M Welding - to Validate or Not? If so, How? ISO 9000, ISO 9001, and ISO 9004 Quality Management Systems Standards 8
V Do we need to Validate a Server/Back-End System? Qualification and Validation (including 21 CFR Part 11) 2
I Using validated SAS JMP to validate Excel for Statistical Analysis? Qualification and Validation (including 21 CFR Part 11) 11
E When to Validate a Process - 21 CFR Part 820.75.(a) 21 CFR Part 820 - US FDA Quality System Regulations (QSR) 8
R How to bond PVC (Tygon) Tubing and Validate the Process ISO 13485:2016 - Medical Device Quality Management Systems 13
T Do I need to validate a basic Excel spreadsheet Quality Assurance and Compliance Software Tools and Solutions 10
R Process Validation - Adhesive Bonding, what to validate and how? Qualification and Validation (including 21 CFR Part 11) 10
N How to Validate a Gage R&R Spread Sheet General Measurement Device and Calibration Topics 3
A How to Validate Effectiveness of Mitigations? ISO 14971 - Medical Device Risk Management 13
C Validate the idea of a Composite Performance Index Quality Tools, Improvement and Analysis 4
M How to determine/validate the "REQUIRED MANPOWER" in an office environment? Service Industry Specific Topics 4
R How to Validate Vision Inspection Systems Inspection, Prints (Drawings), Testing, Sampling and Related Topics 4
C Third-party lab needed to validate cleaning & disinfection? EU Medical Device Regulations 12
Proud Liberal How to validate the geometry of a 8 mm pitch, 22 tooth Gates HTD sprocket Inspection, Prints (Drawings), Testing, Sampling and Related Topics 4
T How does a tool-shop validate molds? Verification vs. Validation ISO 9000, ISO 9001, and ISO 9004 Quality Management Systems Standards 5
D Validate 99.5% reliability 90% confidence - Sample size calculation Reliability Analysis - Predictions, Testing and Standards 4
J Protocol to validate the use of Electronic Records like DHRs, NCMRs, Validations ISO 13485:2016 - Medical Device Quality Management Systems 3
N Periodically validate test reports for raw material - AS9100 7.4.3 AS9100, IAQG, NADCAP and Aerospace related Standards and Requirements 15
V Calibration of a Digital Vernier - Need to check/calibrate/validate all 3 functions? General Measurement Device and Calibration Topics 9
P Last-off-part comparison used to validate the set-up (TS16949 Clause 7.5.1.3) IATF 16949 - Automotive Quality Systems Standard 3
E How to validate product if gage is determined out of calibration Document Control Systems, Procedures, Forms and Templates 19
M Digital Filling System - What do I need to Validate Here? Software Quality Assurance 1
M Seat Cover Product Validation for PPAP - How can I validate my product? APQP and PPAP 12
N Do we need to Validate Outsouced Processes? ISO13485 ISO 13485:2016 - Medical Device Quality Management Systems 12
E GMW3172 - Thermal Shock and PTC cycles required to validate an electronic system Reliability Analysis - Predictions, Testing and Standards 2
A Software Observation during ISO 9001 Audit - Validate and Re-validate Software ISO 9000, ISO 9001, and ISO 9004 Quality Management Systems Standards 13

Similar threads

Top Bottom